Why nearly every tradesperson has a five-star rating (and how to actually choose)

You have three roofers open in three tabs. One is rated 4.8, one 4.9, one a flat 5.0. You keep hunting for the tiebreaker, rereading the same reviews and squinting at decimals, and somewhere around the third pass you start to suspect the whole system is rigged. Nobody is allowed to be this good.

Your suspicion is half right. We analysed 944,548 review scores across four major UK review platforms, and more than four in five of them, 83.5% to be exact, are a perfect five out of five. About one in a hundred is a two or lower. And yet we found no evidence of fraud in those numbers. What we found is a rating system that has stopped being able to rank anyone.

Five stars is a pass mark, not a lie

The pile-up at five stars has a boring explanation. Most reviews are invited just after a job finishes, and most jobs finish fine. A customer whose boiler works again, or whose extension looks like the drawings, gets asked "how did it go?" at the precise moment the answer is "great, thanks". People who abandon a job halfway, or never hire anyone at all, rarely leave a score. What survives into the public record is overwhelmingly completed work, and completed work mostly earns a five.

So a five-star rating is real, and it is worth something. It tells you a tradesperson finishes jobs and leaves customers happy enough to say so. What it cannot tell you is which of three five-star tradespeople is best. The stars have not lied to you. They have stopped being a ranking and turned into a pass mark.

Why we still show star ratings

If stars cannot rank anyone, why do we put them on every profile? Because they still do their first job well: filtering. A solid rating over a real body of reviews screens out the small minority who reliably disappoint. It just cannot separate the finalists, which is why every profile also shows the evidence beyond the stars: credentials checked against official registers, company records in good order, how quickly a business responds, and whether its story holds up across more than one independent source. The stars get a tradesperson onto your shortlist. The rest of the evidence picks the winner.

The three-step way to choose when everyone has five stars

Step 1: Let the stars do their one job, then retire them

Treat anything around 4.5 or above as a pass, and stop reading the decimal. The gap between a 4.8 and a 5.0 is usually one grumpy review and a smaller sample, not a difference in workmanship. Where the stars do still have something to say is at the bottom: a rating below four across a meaningful number of reviews is the rare, genuine warning. Two ratings in the high fours are telling you the same thing, which is that both pass.

Step 2: Read the review count against the trade, not against online shopping

Online shopping trained us to expect thousands of reviews, so thirty can look thin and four can look alarming. For tradespeople that instinct misleads. How many reviews a business collects depends mostly on its trade: how many jobs it does in a year, and whether its customers are homeowners at all. A window company fitting hundreds of quick installations a year piles up reviews at a completely different rate from a builder doing eight extensions.

If a bricklayer you are considering has four reviews, that is not a warning sign. Four is the normal amount of public feedback a bricklaying business generates, the same way six is normal for an architect and seventeen for a builder. Bricklayers do most of their work for builders rather than homeowners, so the public review trail is thin across the whole trade. Judge the four reviews they have, not the eighty they were never going to get.

So calibrate to the trade. A plumber with sixty reviews is well above the typical thirty-four, an electrician with twenty-five is exactly normal, and a roofer with ten is a little light but unremarkable. And once a count sits above the norm for the trade, the reviews stop being a pass mark and start being a track record. That is when reading a handful of recent ones, especially any that mention your kind of job, actually pays off.

Step 3: Treat a second site as a bonus, not a requirement

Most honest trade businesses put their energy into one review site. About five in six of the UK businesses we can see with reviews appear on just one, and that is normal, not a mark against anyone. When a business does show up with reviews on two or more sites (about one in six do, and about one in forty-three appear on three or more), treat it as a bonus worth noticing: two customer bases, invited through different systems, telling the same story. No single page, however good, can give you that kind of confirmation on its own.

A worked example: 5.0 from 3 reviews vs 4.8 from 60

Say you are choosing between two plumbers. One has a perfect 5.0 from 3 reviews; the other a 4.8 from 60. The old instinct says the 5.0 wins, because it is the bigger number. The three steps say otherwise. Step one: both sit comfortably above 4.5, so both pass, and the 0.2 gap carries no information. Step two: the typical plumber with reviews has thirty-four, so sixty is a real track record, while three tells you the business is new to this site or does most of its work off it. That is not disqualifying, just less evidence. Step three: check whether either shows up somewhere else too, and count it in their favour if they do.

On the review evidence alone, the 4.8-from-60 plumber is the safer bet. Sixty near-perfect scores are harder to earn than three perfect ones. The 5.0-from-3 plumber may still be excellent; you simply know less. If everything else about them checks out (the right registrations, a company in good order, a quick reply to your enquiry), the thin review file should not put you off. The point of the method is that the decimal never gets to make the decision.

How we read this for you

This is the same logic our directory applies, at national scale. We collect reviews from multiple independent sources, cross-check them against each other and against official registers and company records, and count independent corroboration in a tradesperson's favour, so the ranking you see already reflects the evidence beyond the stars. Every number on a profile links back to where it came from, and you can check any of it yourself.